Deriving Joint Space Maps of Bundle Compositions and Market Segments: An Application to New Product Options

Marketing Science Institute Technical Working Paper, Report No. 96-122

Posted: 8 Jun 2016

See all articles by Wayne S. DeSarbo

Wayne S. DeSarbo

Pennsylvania State University

Venkatram Ramaswamy

University of Michigan, Stephen M. Ross School of Business

Michel Wedel

University of Maryland - Robert H. Smith School of Business, Marketing Department; University of Groningen - Faculty of Economics and Business

Tammo H.A. Bijmolt

University of Groningen - Department of Marketing & Marketing Research

Date Written: 1996

Abstract

Marketing managers must often anticipate demand for new products and services that do not yet exist in the marketplace. The spatial interaction model described in this paper addresses this challenge by enabling marketers to not only analyze purchase intentions, but also define both market segments and the specific bundles of new product/service options those segments will likely select.

A distinctive feature of this latent structure, multidimensional scaling procedure is that it simultaneously extracts market segments and positions the product options on a joint space map. The closer a product option is to a particular segment, the higher the likelihood of its being chosen by that segment. The threshold utility for each segment is used to define the bundle of preferred product options. Estimates of the probability of each consumer belonging to the derived segments are also obtained simultaneously.

In addition, the locations of the product/service options on the map can be directly reparameterized as a function of specific product characteristics-such as physical aspects, functionality, or price-in order to aid managers in positioning these options in the marketplace. Moreover, if additional background data on consumers are available, this information can be related to the segment locations to help refine the targeting of specific markets.

Application to New Product Options To illustrate the use of our model, the authors provide an actual commercial application using pick-any data for several advanced product options. A major high-tech firm conducted extensive consumer interviews to investigate the desirability of some 23 product options that it wanted to "pre-market" to automobile manufacturers. The main objectives of the study were to understand how consumers judged the complementarity among the various options, and to define both market segments and the bundles (subsets) of options that each segment preferred.

In personal interviews with a sample of 376 consumers, moderators described the functions, features, and benefits of each of the options, along with their costs. Respondents were asked to choose those options they would definitely consider purchasing as a bundle with their next vehicle. The survey also included questions about vehicle preferences, ownership history, driving habits, and personal background.

The results point to three market segments with three distinct, though overlapping, bundle choices. That is, Segment 1's option preferences are nested within those of Segment 2, and Segment 2's preferences are nested within those of Segment 3. The common (primary) bundle consists of five options-visibility lighting, all weather windows, fast brake lights, wide view windows, and self-sealing tires-that appear to offer the most utility for the added price. When considered with (unreported) background information, the results suggest that the primary bundle could be offered on certain vehicles as standard equipment, the secondary bundle as a low-end package, and the tertiary bundle as a high-end package.

Additional Applications Although the new product option example discussed here employs two-way, pick-any data, the model also accommodates three-way data collected over different scenarios or usage situations. In addition, it allows the dimensions for each situation/scenario to have a different impact within each segment.

And, in addition to consumer judgments, the spatial interaction model can be used to analyze consumer behavior, e.g.,

- to investigate complementarity among different types of products over time,

- to represent the joint purchase of various brands within the same product class over time,

- to examine choice or purchase shifts as a result of various treatments, and

- to spatially represent the results of choice-based conjoint studies.

Suggested Citation

DeSarbo, Wayne S. and Ramaswamy, Venkatram and Wedel, Michel and Bijmolt, Tammo H.A., Deriving Joint Space Maps of Bundle Compositions and Market Segments: An Application to New Product Options (1996). Marketing Science Institute Technical Working Paper, Report No. 96-122. Available at SSRN: https://ssrn.com/abstract=2791146

Wayne S. DeSarbo (Contact Author)

Pennsylvania State University ( email )

University Park
State College, PA 16802
United States

Venkatram Ramaswamy

University of Michigan, Stephen M. Ross School of Business ( email )

701 Tappan Street
Ann Arbor, MI MI 48109-1234
United States
734-763-5932 (Phone)
734-936-0279 (Fax)

Michel Wedel

University of Maryland - Robert H. Smith School of Business, Marketing Department ( email )

College Park, MD 20742
United States
301.405.2162 (Phone)
301.405.0146 (Fax)

HOME PAGE: http://www.rhsmith.umd.edu/marketing/faculty/wedel.html

University of Groningen - Faculty of Economics and Business ( email )

Postbus 72
9700 AB Groningen
Netherlands

Tammo H.A. Bijmolt

University of Groningen - Department of Marketing & Marketing Research ( email )

Nettelbosje 2
Groningen, 9747 AE
Netherlands

Register to save articles to
your library

Register

Paper statistics

Abstract Views
141
PlumX Metrics